First published online July 23, 2007
British Journal of Radiology (2007) 80, 532-536
© 2007 British Institute of Radiology
doi: 10.1259/bjr/84718521
Effect of age and gender on the number and distribution of sites in Paget's disease of bone
M J Haddaway, BSc1,
M W J Davie, FRCP2,
I W MCCall, FRCR1 and
S Howdle, FRCR3
1 Department of Diagnostic Imaging and, 2 Charles Salt Centre for Human Metabolism, Robert Jones & Agnes Hunt Orthopaedic Hospital, Oswestry, Shropshire SY10 7AG, 3 X-ray Department, Princess Royal Hospital, Telford, Shropshire, UK
Correspondence: Mr Mike Haddaway, Diagnostic Imaging Department, Robert Jones & Agnes Hunt Orthopaedic Hospital, Twmpath, Oswestry, Shropshire SY10 7AG, UK. E-mail: mike.haddaway{at}rjah.nhs.uk
 |
Abstract
|
|---|
Paget's disease of bone (PDB) is reportedly declining in prevalence and severity, with increasing numbers of monostotic cases. Some accounts suggest that these findings are more evident in women, and that monostotic disease is unexpectedly frequent at certain sites. We have studied whether birth date or gender is associated with the number of sites affected and with the distribution of sites in monostotic disease and, by reviewing 100 follow-up 99Tcm methylene diphosphonate (MDP) scans, whether additional sites appear after initial diagnosis. Scintigraphic scans from 171 male (age 40–91 years) and 179 female (44–88 years) consecutive referrals with PDB were reviewed. Patients were analysed by referral date (1982–1992 and 1993–2001), and by their median date of birth (before (PRE21) and after (POST21) 1921). Mean age of pre-1993 patients was 69 years and 75 years for referrals after 1993. Younger patients had more monostotic disease (POST21 vs PRE21 subjects (47% vs 28%)), with a more marked trend in females (52% vs 25%), and POST21 females had fewer polyostotic sites than males (p<0.05), whereas the number in PRE21 males and females was similar. Monostotic females, but not males, showed an excess of tibial involvement. The spine was less involved in monostotic disease. Follow-up scans in 100 patients revealed no new sites. The incidence of monostotic disease has doubled over the last 30 years, but diminishing site involvement appears to be more marked in females. The lesser involvement at the axial sites in monostotic disease may lead to overestimation of the decline in PDB based on abdominal radiographs.
|
Introduction
|
|---|
In many countries, particularly those with higher incidence rates of Paget's disease of bone, the frequency of the condition is declining [1–7], although in Italy, where the prevalence is lower, radiological assessment of pelvic radiographs (including the lumbar spine) showed a slight increase between 1986/7 and 2001/2, especially among patients over 75 years of age [8]. The disease also appears to be showing diminishing severity as the proportion of patients with bone disease at multiple sites falls, and lower values of total serum alkaline phosphatase (total ALP) are found at initial presentation [9]. The number of sites involved at presentation in men compared with women has variously been described as increased [10] or no different [11], while a lower proportion of skeletal involvement in men was described by Harinck et al [12]. There is also an indication from Dunedin that the decline in the severity of Paget's disease may be more marked in women [1].
Over the last 40 years in New Zealand, the decline in the severity of Paget's disease and the number of anatomical sites has been mirrored by an increase in the proportion of monostotic disease from 24% to 36% [1]. The distribution of Paget's disease in the skeleton may also be affected by the decline in the number of sites involved. Monostotic disease may be characterized by a higher than expected incidence in the tibia [11, 12], in excess of that expected from the polyostotic distribution [11]. Women are also thought to have a greater incidence of monostotic disease than men, although whether women have a tibial excess remains uncertain. Descriptions of gender differences in the distribution of Paget's disease have been confined to polyostotic and monostotic cases combined [11], and might be more evident in a study of monostotic cases of Paget's disease. The anatomical distribution of Paget's disease in the skeleton has also acquired greater significance in recent years because of large studies reporting prevalence rates based on the presence of Paget's disease in abdominal radiographs [5]. If more cases of Paget's disease have only peripheral sites involved, studies of prevalence based on abdominal radiographs may give falsely low rates of Paget's disease. The prevalence ratio of men: women in referred studies has tended to be lower (1.2–1.9) than in screening studies (based on abdominal radiographs), which tend to show a higher male predominance (
2.0) [5, 7], although this may show a decline with time [5, 8]. Moreover, the increase in the proportion of patients with monostotic disease in Europe from 24% in 1984 [13] to as high as 40% in Italy [8] may also lead to underestimation of the prevalence in both sexes.
We have examined scintiscans from patients referred over 20 years to investigate whether there were any differences in the number of sites involved and in the proportion of patients with monostotic disease according to age and gender. We have also investigated whether there is any gender difference in the bones involved and whether changes in the skeletal distribution of Paget's disease might affect the estimation of prevalence based on abdominal radiography. As it has been reported that new sites can appear at any time, converting monostotic cases to polyostotic cases [14], we have also investigated whether new sites have appeared in scans followed up some years later.
|
Methods and materials
|
|---|
The bone scintigraphy of 350 consecutive patients, referred to this institution between 1982 and 2001 (171 males, 179 females), was reviewed. The protocol is for all patients suspected of having Paget's disease to undergo bone scintigraphy. The diagnosis of Paget's disease had been established from a combination of clinical findings, radiology and biochemistry. All patients were scanned for routine clinical evaluation and, initially, treatment response was assessed, for clinical purposes, by serial imaging. Of these patients reassessed in this way, 100 were selected at random to identify the presence of new sites of disease (SH and IWM). All scans were retrospectively reassessed independently by consensus (IWM and MJH), and the sites of Paget's disease involvement were recorded. All authors who were involved in either assessment were blinded to previous evaluations of the scan images.
Whole-body bone scintigraphy was performed after 3 h, following the intravenous injection of 600 MBq of 99Tcm methylene diphosphonate (99Tcm MDP). Anterior and posterior views were acquired to include the whole skeleton, except in those cases (rare) where the patient width exceeded the field of view of the camera. Images were presented at two different levels of intensity to allow for the wide latitude of activity levels present in bone scintigraphy of patients with Paget's disease. Scans were performed on one of two gamma cameras. Prior to 1990, patients were scanned on a Scintronix Digicamera (Scintronix, Livingston, UK) and, subsequently, they were scanned on a Toshiba GCA901/A gamma camera (Toshiba, Tokyo, Japan). Effective dose from a 99Tcm MDP bone scan is of the order of 3 mSv.
The scans were analysed visually and the sites of involvement with Paget's disease recorded, the principle features being increased uptake and bony expansion. Uptake was defined as increased relative to the remainder of the skeleton. Cross-reference with plain radiographs was made in almost all cases and whenever there was doubt about a site. Normal bone from the same patient was used as a standard for comparison with areas of disease.
Although all sites affected by Paget's disease were recorded, the vertebrae were divided into three spinal areas (dorsal, lumbar or cervical) in order to compare with other reports. If more than one vertebra was affected in one spinal area, but no sites appeared elsewhere in the skeleton, this was counted as monostotic [11]. Appropriate comment is made in the results if counting of individual vertebrae would alter the data or conclusions. Bilateral uptake in the same bone on the right and left may also be regarded as monostotic or polyostotic. In the present study, the pelvis was regarded as one bone, irrespective of whether one or both sides of the pelvis were involved, and bilateral uptake was taken as a single site. No patient with monostotic disease had bilateral sites in the upper or lower limbs. An indication is given in the results about how counting each bone separately (except the pelvis as noted above) would alter the number of sites.
Statistics
Data were analysed using SPSS v14. Descriptive statistics were expressed as mean ± SD. Student's t-test was used to compare ages between groups. 
2 and Fisher's exact test were employed for comparison between age and time of presentation groups. Differences between the numbers of sites affected by Paget's disease were tested using the Mann–Whitney U-test.
|
Results
|
|---|
Between 1982 and 2001, a diagnosis of Paget's disease of bone was made in 350 patients (171 males, 179 females). Age range was 40–91 years in males and 44–88 years in females. The distribution of patients by gender and age at presentation in the two decades appears in Table 1
. Men were younger than women in the first decade (p<0.01) but, by the second decade, mean age at referral had advanced by 6 years (Table 1) and the gender difference had disappeared.
The number of sites at referral did not differ between the two decades, but men had more sites on average than women overall (p<0.05), and men had more sites affected than women before 1993 (p<0.05) (Table 1
). Individual scoring of all sites/bones yielded a greater average number of sites in polyostotic subjects, with males having 6.33 and females 4.77 sites per subject (5.5 if the genders are combined). The number of sites in males is still significantly greater than for females (p<0.05). Monostotic disease affected 64 women and 64 men, compared with 115 women and 107 men with polyostotic disease, but the proportion of patients with monostotic disease did not change between 1982–92 and 1993–2001 (Table 1).
The prevalence of Paget's disease was also investigated according to the subject's date of birth. The median date of birth was May 1921. Subjects were split into cohorts born before (PRE21) and after (POST21) this date (Table 2). Monostotic disease was present in a greater proportion of POST21 subjects (80/174) than PRE21 (48/176) (p<0.01). Only in one patient, classified as monostotic, would the classification be altered if individual bones were counted. This patient had sites affected by Paget's disease at L3, L4 and L5 vertebrae. This would therefore have altered the monostotic to polyostotic ratio by one female subject.
View this table:
[in this window]
[in a new window]
|
Table 2. Summary of data from subjects who were born prior to May 1921(median date of birth) and those born after May 1921
|
|
When divided by gender, there was no difference in the prevalence of monostotic vs polyostotic disease in men and women overall. The data in Table 2 indicated that monostotic disease might have become more frequent in females than in males after 1921 and that the proportion of patients with monostotic disease had increased to a greater extent in females (25% to 52%) than in males (31% to 42%) from PRE21 to POST21. However, the increase in monostotic disease in females was only a trend, achieving significance at no more than the 10% level.
In contrast with the increasing number of patients with only one site affected, patients with polyostotic disease born after 1921 did not show a decline in the number of skeletal sites affected compared with those born before 1921. There was a tendency (although not significant) for POST21 men to show an increase in the number of sites affected in polyostotic disease compared with PRE21 men. POST21 men with polyostotic disease did, however, have more sites affected in the skeleton than POST21 women (p<0.05; Table 2).
Differing involvement of particular skeletal sites existed when analysed either by gender or in relation to the presence of monostotic or polyostotic disease. There was an excess of tibial involvement in females (17/64) compared with males (6/64; p<0.05) in patients with monostotic disease, but this was not the case in polyostotic disease. Furthermore, patients (sexes combined) with monostotic disease were less likely to demonstrate Paget's disease in the spine (7/128) than those with polyostotic disease (146/222; p<0.001), but this did not differ between males and females.
As some surveys have described the epidemiology of Paget's disease from the prevalence of abnormal bone visible in radiographs of the abdomen [5, 15], we examined how many patients would be overlooked by this technique compared with 99Tcm MDP scanning. For all patients, 15% would be overlooked through not having Paget's disease in the abdominal field (lumbar spine, sacrum and pelvis). For monostotic disease, the proportion would be higher with 20% of men and 39% of women being excluded.
To investigate whether additional sites appeared after the initial scans, follow-up scans of 100 patients were evaluated (see Methods). This set, in which 42% of patients had monostotic disease, was aged 37–89 years (mean 66.5 years) and included 49 males and 51 females. These patients were initially scanned between 1983 and 2000, and had subsequent scans between 1 and 18 years after the initial scan (mean 5.5 years; median 4 years). No additional sites were detected.
|
Discussion
|
|---|
The age at presentation in our series is increasing (Table 1) consistent with the findings in New Zealand [1]. The number of sites at presentation did not change between the ninth and tenth decades of the twentieth century (Table 1) nor between those born before or after 1921 (Table 2). There was however a trend towards lower values in women born after 1921 (Table 2), and women born after this date had significantly fewer sites than men in the same era. Men did not display the same downward trend in sites with later birth. Such a gender-dependent decline in polyostotic sites also appears in a series of 70- to 79-year-old females collected from Dunedin between 1983 and 1990 [7]. Later studies revealed an inverse relationship between the number of sites and the year of birth [1], but made no comment on gender specificity. Other studies from Europe between 1986 and 1991 [11, 12] report that the number of sites involved in polyostotic disease is between 6.4 (calculated from the data of Harinck et al [12]) and 6.0 in the Sheffield analysis [11].
The trend towards fewer sites is echoed in the rise in the number of patients with monostotic disease. The proportion of 37% in the present study is consistent with the increasing proportion of patients with monostotic disease. This phenomenon is depicted in Figure 1 showing the trend in large reported series of patients with Paget's disease, assessed by scintiscans.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 1. Graphical representation of the proportion of patients with monostotic disease in patient series investigated by scintigraphy according to year of publication of the data. Note the increase in patients with only one site of Paget's disease of bone. References: Shirazi et al, 1974 [16]; Kanis, 1991 [11]; Meunier et al, 1987 [10]; Salson, 1981 [as quoted in Kanis [11]); Cundy et al, 2004 [1]; Vellenga et al, 1984 [13]; Wellman et al, 1977 [17]; Monfort et al, 1999 [18]; Harinck et al, 1986 [12]; Alvarez et al, 1997 [19]; Morales-Piga et al, 2002 [6]; Merrick and Merrick, 1985 [20]; Gennari et al, 2005 [8]. The data from Morales-Piga et al (2002) [6] represent an average of two groups described in the paper (94/233). The data from Cundy et al (2004) [1] is an average of the percentage with monostotic disease between 1985–1993 (24%) and 1994–2002 (36%).
|
|
The proportion of patients with monostotic disease is higher in series using abdominal radiography [17]. Reports using abdominal radiographs are not included in Figure 1 as scintiscans and abdominal radiographs do not compare the same skeletal sites.
The increasing proportion of patients with monostotic disease over the last 30 years (Figure 1) mirrors the present data showing an increase in the proportion of patients with monostotic disease after 1921. The change is more obvious among females, although in our data the change in females alone only reaches significance at the 10% level. As there is a trend towards more patients having only one site involved, we investigated whether the number of polyostotic sites was showing a decline. However, this was not the case (Table 2), either for the whole group or for individual genders.
Gender dependency was evident in the sites affected by monostotic disease. Earlier studies had shown that patients with monostotic distribution had a lower prevalence of disease in the spine, and our analysis confirmed the low incidence in the spine in both sexes [11]. Other studies had found that the tibia was more likely to be affected in patients with monostotic disease [11, 20]. In the present study, 36% of patients with monostotic disease had Paget's disease in the tibia, but almost three-quarters of these were female.
The trend towards fewer affected sites in patients with Paget's disease born later and some of the site specificity is largely a female phenomenon in our series of patients. The reason for this finding is not clear. Exposure to dogs has been suggested as a factor involved in the appearance of Paget's disease of bone [21]. We have not quantified the exposure to dogs to identify whether women are less exposed than men. We have not yet elucidated whether female subjects are more likely to have a family history, or whether localization of Paget's disease might have an inherited element or be associated with specific haplotypes. Such questions will be properly addressed by the ongoing PRISM study [22].
The trend towards fewer sites and to more monostotic disease, which, particularly in women, may be peripheral, will bias conclusions made about the prevalence of Paget's disease derived from abdominal radiographs [5, 23]. The prevalence of Paget's disease using abdominal radiographs will be underestimated. This may be particularly relevant in women in whom the increase in monostotic disease is associated with fewer sites within the abdominal field (spine) and more outside (tibia).
Apart from the difficulties in defining monostotic disease (see Methods), it is also possible that monostotic disease in younger patients at presentation might become polyostotic if new sites appeared. The emergence of six new sites was reported in one series of patients with Paget's disease, but it is not clear whether this refers to new sites appearing remote from the initial presentation or to the reappearance of existing sites after treatment [24]. We were unable to identify any new sites appearing over a long period of time. At the present time, with effective treatments for Paget's disease being available, it is unlikely that new sites will become apparent, and our data indicate that the appearance of new sites is only a remote possibility.
Paget's disease of bone appears to be declining in severity, as evidenced by increasing monostotic disease and fewer polyostotic sites, but this is most evident in women. A decline in prevalence as ascertained by abdominal radiographs may be overestimated as the disease tends to affect both fewer bones and more peripheral sites.
Received for publication July 19, 2006.
Revision received October 23, 2006.
Accepted for publication November 7, 2006.
|
References
|
|---|
- Cundy HR, Gamble G, Wattie D, Rutland M, Cundy T. Paget's disease of bone in New Zealand: continued decline in disease severity. Calcif Tissue Int 2004;75:358–64.[CrossRef][Medline]
- Barker DJ, Chamberlain AT, Guyer PB, Gardner MJ. Paget's disease of bone: the Lancashire focus. Br Med J 1980;280:1105–7.[Abstract/Free Full Text]
- Barker DJ, Clough PW, Guyer PB, Gardner MJ. Paget's disease of bone in 14 British towns. Br Med J 1977;1:1181–3.[Abstract/Free Full Text]
- van Staa TP, Selby P, Leufkens HG, Lyles K, Sprafka JM, Cooper C. Incidence and natural history of Paget's disease of bone in England and Wales. J Bone Miner Res 2002;17:465–71.[CrossRef][Medline]
- Cooper C, Schafheutle K, Dennison E, Kellingray S, Guyer P, Barker D. The epidemiology of Paget's disease in Britain: is the prevalence decreasing? J Bone Miner Res 1999;14:192–7.[CrossRef][Medline]
- Morales-Piga AA, Bachiller-Corral FJ, Abraira V, Beltran J, Rapado A. Is clinical expressiveness of Paget's disease of bone decreasing? Bone 2002;30:399–403.[Medline]
- Doyle T, Gunn J, Anderson G, Gill M, Cundy T. Paget's disease in New Zealand: evidence for declining prevalence. Bone 2002;31:616–19.[Medline]
- Gennari L, Di Stefano M, Merlotti D, Giordano N, Martini G, Tamone C, et al. Prevalence of Paget's disease of bone in Italy. J Bone Miner Res 2005;20:1845–50.[CrossRef][Medline]
- Cundy T, McAnulty K, Wattie D, Gamble G, Rutland M, Ibbertson HK. Evidence for secular change in Paget's disease. Bone 1997;20:69–71.[Medline]
- Meunier PJ, Salson C, Mathieu L, Chapuy MC, Delmas P, Alexandre C, et al. Skeletal distribution and biochemical parameters of Paget's disease. Clin Orthop Relat Res 1987;217:37–44.[Medline]
- Kanis JA, editor. Pathophysiology and treatment of Paget's dDisease of bone (Chapter 3). London, UK: Dunitz, 1991
- Harinck HI, Bijvoet OL, Vellenga CJ, Blanksma HJ, Frijlink WB. Relation between signs and symptoms in Paget's disease of bone. Q J Med 1986;58:133–51.[Medline]
- Vellenga CJ, Pauwels EK, Bijvoet OL, Frijlink WB, Mulder JD, Hermans J. Untreated Paget disease of bone studied by scintigraphy. Radiology 1984;153:799–805.[Abstract/Free Full Text]
- Mirra JM, Brien EW, Tehranzadeh J. Paget's disease of bone: review with emphasis on radiologic features, Part I. Skeletal Radiol 1995;24:163–71.[Medline]
- Guyer PB. Paget's disease of bone: the anatomical distribution. Metab Bone Dis Relat Res 1981;3:239–41.[CrossRef][Medline]
- Shirazi PH, Ryan WG, Fordham EW. Bone scanning in evaluation of Paget's disease of bone. CRC Crit Rev Clin Radiol Nucl Med 1974;5:523–58.[Medline]
- Wellman HN, Schauwecker D, Robb JA, Khairi MR, Johnston CC. Skeletal scintimaging and radiography in the diagnosis and management of Paget's disease. Clin Orthop Relat Res 1977;127:55–62.[Medline]
- Monfort J, Sala DR, Romero AB, Duro JC, Maymo J, Carbonell J. Epidemiological, clinical, biochemical, and imaging characteristics of monostotic and polyostotic Paget's disease. Bone 1999;24(5 Suppl):13S–14S.
- Alvarez L, Peris P, Pons F, GuaNabens N, Herranz R, Monegal A, et al. Relationship between biochemical markers of bone turnover and bone scintigraphic indices in assessment of Paget's disease activity. Arthritis Rheum 1997;40:461–8.[Medline]
- Merrick MV, Merrick JM. Observations on the natural history of Paget's disease. Clin Radiol 1985;36:169–74.[CrossRef][Medline]
- O'Driscoll JB, Anderson DC. Past pets and Paget's disease. Lancet 1985;2:919–21.[Medline]
- Langston AL, McCallum M, Campbell MK, Robertson C, Ralston SH. An integrated approach to consumer representation and involvement in a multicentre randomized controlled trial. Clin Trials 2005;2:80–7.[Abstract/Free Full Text]
- Guyer PB, Clough PW. Paget's diseases of bone: some observations on the relation of the skeletal distribution to pathogenesis. Clin Radiol 1978;29:421–6.[CrossRef][Medline]
- Vellenga CJ, Pauwels EK, Bijvoet OL, Frijlink WB. Scintigraphic aspects of the recurrence of treated Paget's disease of bone. J Nucl Med 1981;22:510–17.[Abstract/Free Full Text]
Top
Abstract
Introduction
